Formalizing Statistical Causality via Modal Logic
This work provides a formal framework for causality in statistics, which is incremental as it builds on existing causal theories like Pearl's do-calculus.
The authors tackled the problem of formalizing statistical causality by proposing Statistical Causality Language (StaCL), a formal language with modal operators for expressing causal effects and inference requirements, and demonstrated that it can derive Pearl's do-calculus rules and explain causal inference correctness.
We propose a formal language for describing and explaining statistical causality. Concretely, we define Statistical Causality Language (StaCL) for expressing causal effects and specifying the requirements for causal inference. StaCL incorporates modal operators for interventions to express causal properties between probability distributions in different possible worlds in a Kripke model. We formalize axioms for probability distributions, interventions, and causal predicates using StaCL formulas. These axioms are expressive enough to derive the rules of Pearl's do-calculus. Finally, we demonstrate by examples that StaCL can be used to specify and explain the correctness of statistical causal inference.